Project abstract Study of movement disorders suggests that dopamine (DA) and the broader nigrostriatal circuit may play a specialized role in self-timed movements, for which the drive to move must be generated internally rather than in reaction to external events. However, classic experiments suggested that DA neurons (DANs) encode reward-prediction errors (RPEs) that occur too late to facilitate movements. In project 1, our collaborators present an updated temporal difference (TD) model for which, under appropriate conditions, RPE/DA signals ?ramp-up? during ongoing behavior. These signals could be associated with, or facilitate, self-timed movements. We will address the hypothesis in three ways, using a self-timed movement task in mice. First, we will record from genetically defined DANs during self-timed movements, to assess the relationship between DAN activity and movement time. We already have strong preliminary evidence that DANs indeed ramp up their activity before self-timed movements, with the slope of ramping inversely related to the movement time. Second, we will test whether DA ramps are causal to self-timed movements, by optogenetically stimulating genetically defined DANs and examining the effect on the timing of self-timed movements. Third, the TD/RPE theory explains how DANs can evince ramping activity, but does not address how DA ramping affects downstream targets. We hypothesize that DANs facilitate self-timed movements by oppositely modulating striatal spiny projection neurons (SPNs) of the direct and indirect striatal pathways. To test this hypothesis, we will simultaneously monitor pairwise activity from genetically identified DANs, dSPNs or iSPNs to assess 1) the relationship of dSPN/iSPN balance and movement time, and 2) the cell types' influence on each other. These experiments will provide crucial information on the function of the key nigrostriatal circuit, grounded in a novel theory that makes testable hypotheses.